Next: , Up: (dir)

Plzip Manual

This manual is for Plzip (version 1.3-pre1, 25 November 2014).


Copyright © 2009-2014 Antonio Diaz Diaz.

This manual is free documentation: you have unlimited permission to copy, distribute and modify it.


Next: , Previous: Top, Up: Top

1 Introduction

Plzip is a massively parallel (multi-threaded) lossless data compressor based on the lzlib compression library, with a user interface similar to the one of lzip, bzip2 or gzip.

Plzip can compress/decompress large files on multiprocessor machines much faster than lzip, at the cost of a slightly reduced compression ratio (0.4 to 2 percent larger compressed files). Note that the number of usable threads is limited by file size; on files larger than a few GB plzip can use hundreds of processors, but on files of only a few MB plzip is no faster than lzip (see Minimum file sizes).

Plzip uses the lzip file format; the files produced by plzip are fully compatible with lzip-1.4 or newer, and can be rescued with lziprecover.

The lzip file format is designed for data sharing and long-term archiving, taking into account both data integrity and decoder availability:

A nice feature of the lzip format is that a corrupt byte is easier to repair the nearer it is from the beginning of the file. Therefore, with the help of lziprecover, losing an entire archive just because of a corrupt byte near the beginning is a thing of the past.

Plzip uses the same well-defined exit status values used by lzip and bzip2, which makes it safer than compressors returning ambiguous warning values (like gzip) when it is used as a back end for other programs like tar or zutils.

Plzip will automatically use the smallest possible dictionary size for each file without exceeding the given limit. Keep in mind that the decompression memory requirement is affected at compression time by the choice of dictionary size limit (see Memory requirements).

When compressing, plzip replaces every file given in the command line with a compressed version of itself, with the name "original_name.lz". When decompressing, plzip attempts to guess the name for the decompressed file from that of the compressed file as follows:

filename.lz becomes filename
filename.tlz becomes filename.tar
anyothername becomes anyothername.out

(De)compressing a file is much like copying or moving it; therefore plzip preserves the access and modification dates, permissions, and, when possible, ownership of the file just as "cp -p" does. (If the user ID or the group ID can't be duplicated, the file permission bits S_ISUID and S_ISGID are cleared).

Plzip is able to read from some types of non regular files if the '--stdout' option is specified.

If no file names are specified, plzip compresses (or decompresses) from standard input to standard output. In this case, plzip will decline to write compressed output to a terminal, as this would be entirely incomprehensible and therefore pointless.

Plzip will correctly decompress a file which is the concatenation of two or more compressed files. The result is the concatenation of the corresponding uncompressed files. Integrity testing of concatenated compressed files is also supported.

WARNING! Even if plzip is bug-free, other causes may result in a corrupt compressed file (bugs in the system libraries, memory errors, etc). Therefore, if the data you are going to compress is important, give the '--keep' option to plzip and do not remove the original file until you verify the compressed file with a command like 'plzip -cd file.lz | cmp file -'.


Next: , Previous: Introduction, Up: Top

2 Program design

When compressing, plzip divides the input file into chunks and compresses as many chunks simultaneously as worker threads are chosen, creating a multi-member compressed file.

When decompressing, plzip decompresses as many members simultaneously as worker threads are chosen. Files that were compressed with lzip will not be decompressed faster than using lzip (unless the '-b' option was used) because lzip usually produces single-member files, which can't be decompressed in parallel.

For each input file, a splitter thread and several worker threads are created, acting the main thread as muxer (multiplexer) thread. A "packet courier" takes care of data transfers among threads and limits the maximum number of data blocks (packets) being processed simultaneously.

The splitter reads data blocks from the input file, and distributes them to the workers. The workers (de)compress the blocks received from the splitter. The muxer collects processed packets from the workers, and writes them to the output file.

When decompressing from a regular file, the splitter is removed and the workers read directly from the input file. If the output file is also a regular file, the muxer is also removed and the workers write directly to the output file. With these optimizations, the use of RAM is greatly reduced and the decompression speed of large files with many members is only limited by the number of processors available and by I/O speed.


Next: , Previous: Program design, Up: Top

3 Invoking plzip

The format for running plzip is:

     plzip [options] [files]

Plzip supports the following options:

'-h'
'--help'
Print an informative help message describing the options and exit.
'-V'
'--version'
Print the version number of plzip on the standard output and exit.


'-B bytes'
'--data-size=bytes'
Set the size of the input data blocks, in bytes. The input file will be divided in chunks of this size before compression is performed. Valid values range from 8 KiB to 1 GiB. Default value is two times the dictionary size. Plzip will reduce the dictionary size if it is larger than the chosen data size.
'-c'
'--stdout'
Compress or decompress to standard output. Needed when reading from a named pipe (fifo) or from a device.
'-d'
'--decompress'
Decompress.
'-f'
'--force'
Force overwrite of output files.
'-F'
'--recompress'
Force recompression of files whose name already has the '.lz' or '.tlz' suffix.
'-k'
'--keep'
Keep (don't delete) input files during compression or decompression.
'-m bytes'
'--match-length=bytes'
Set the match length limit in bytes. After a match this long is found, the search is finished. Valid values range from 5 to 273. Larger values usually give better compression ratios but longer compression times.
'-n n'
'--threads=n'
Set the number of worker threads. Valid values range from 1 to "as many as your system can support". If this option is not used, plzip tries to detect the number of processors in the system and use it as default value. 'plzip --help' shows the system's default value.

Note that the number of usable threads is limited to ceil( file_size / data_size ) during compression (see Minimum file sizes), and to the number of members in the input during decompression.

'-o file'
'--output=file'
When reading from standard input and '--stdout' has not been specified, use 'file' as the virtual name of the uncompressed file. This produces a file named 'file' when decompressing, and a file named 'file.lz' when compressing.
'-q'
'--quiet'
Quiet operation. Suppress all messages.
'-s bytes'
'--dictionary-size=bytes'
Set the dictionary size limit in bytes. Valid values range from 4 KiB to 512 MiB. Plzip will use the smallest possible dictionary size for each file without exceeding this limit. Note that dictionary sizes are quantized. If the specified size does not match one of the valid sizes, it will be rounded upwards by adding up to (bytes / 16) to it.

For maximum compression you should use a dictionary size limit as large as possible, but keep in mind that the decompression memory requirement is affected at compression time by the choice of dictionary size limit.

'-t'
'--test'
Check integrity of the specified file(s), but don't decompress them. This really performs a trial decompression and throws away the result. Use it together with '-v' to see information about the file.
'-v'
'--verbose'
Verbose mode.
When compressing, show the compression ratio for each file processed. A second '-v' shows the progress of compression.
When decompressing or testing, further -v's (up to 4) increase the verbosity level, showing status, compression ratio, dictionary size, decompressed size, and compressed size.
'-1 .. -9'
Set the compression parameters (dictionary size and match length limit) as shown in the table below. Note that '-9' can be much slower than '-1'. These options have no effect when decompressing.

The bidimensional parameter space of LZMA can't be mapped to a linear scale optimal for all files. If your files are large, very repetitive, etc, you may need to use the '--match-length' and '--dictionary-size' options directly to achieve optimal performance.

Level Dictionary size Match length limit
-1 1 MiB 5 bytes
-2 1.5 MiB 6 bytes
-3 2 MiB 8 bytes
-4 3 MiB 12 bytes
-5 4 MiB 20 bytes
-6 8 MiB 36 bytes
-7 16 MiB 68 bytes
-8 24 MiB 132 bytes
-9 32 MiB 273 bytes

'--fast'
'--best'
Aliases for GNU gzip compatibility.

Numbers given as arguments to options may be followed by a multiplier and an optional 'B' for "byte".

Table of SI and binary prefixes (unit multipliers):

Prefix Value | Prefix Value
k kilobyte (10^3 = 1000) | Ki kibibyte (2^10 = 1024)
M megabyte (10^6) | Mi mebibyte (2^20)
G gigabyte (10^9) | Gi gibibyte (2^30)
T terabyte (10^12) | Ti tebibyte (2^40)
P petabyte (10^15) | Pi pebibyte (2^50)
E exabyte (10^18) | Ei exbibyte (2^60)
Z zettabyte (10^21) | Zi zebibyte (2^70)
Y yottabyte (10^24) | Yi yobibyte (2^80)


Exit status: 0 for a normal exit, 1 for environmental problems (file not found, invalid flags, I/O errors, etc), 2 to indicate a corrupt or invalid input file, 3 for an internal consistency error (eg, bug) which caused plzip to panic.


Next: , Previous: Invoking plzip, Up: Top

4 File format

Perfection is reached, not when there is no longer anything to add, but when there is no longer anything to take away.
-- Antoine de Saint-Exupery


In the diagram below, a box like this:
+---+
|   | <-- the vertical bars might be missing
+---+

represents one byte; a box like this:

+==============+
|              |
+==============+

represents a variable number of bytes.


A lzip file consists of a series of "members" (compressed data sets). The members simply appear one after another in the file, with no additional information before, between, or after them.

Each member has the following structure:

+--+--+--+--+----+----+=============+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| ID string | VN | DS | Lzma stream | CRC32 |   Data size   |  Member size  |
+--+--+--+--+----+----+=============+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

All multibyte values are stored in little endian order.

'ID string'
A four byte string, identifying the lzip format, with the value "LZIP" (0x4C, 0x5A, 0x49, 0x50).
'VN (version number, 1 byte)'
Just in case something needs to be modified in the future. 1 for now.
'DS (coded dictionary size, 1 byte)'
Lzip divides the distance between any two powers of 2 into 8 equally spaced intervals, named "wedges". The dictionary size is calculated by taking a power of 2 (the base size) and substracting from it a number of wedges between 0 and 7. The size of a wedge is (base_size / 16).
Bits 4-0 contain the base 2 logarithm of the base size (12 to 29).
Bits 7-5 contain the number of wedges (0 to 7) to substract from the base size to obtain the dictionary size.
Example: 0xD3 = 2^19 - 6 * 2^15 = 512 KiB - 6 * 32 KiB = 320 KiB
Valid values for dictionary size range from 4 KiB to 512 MiB.
'Lzma stream'
The lzma stream, finished by an end of stream marker. Uses default values for encoder properties. See the lzip manual for a full description.
'CRC32 (4 bytes)'
CRC of the uncompressed original data.
'Data size (8 bytes)'
Size of the uncompressed original data.
'Member size (8 bytes)'
Total size of the member, including header and trailer. This field acts as a distributed index, allows the verification of stream integrity, and facilitates safe recovery of undamaged members from multi-member files.


Next: , Previous: File format, Up: Top

5 Memory required to compress and decompress

The amount of memory required per thread is approximately the following:


Next: , Previous: Memory requirements, Up: Top

6 Minimum file sizes required for full compression speed

When compressing, plzip divides the input file into chunks and compresses as many chunks simultaneously as worker threads are chosen, creating a multi-member compressed file.

For this to work as expected (and roughly multiply the compression speed by the number of available processors), the uncompressed file must be at least as large as the number of worker threads times the chunk size (see -data-size). Else some processors will not get any data to compress, and compression will be proportionally slower. The maximum speed increase achievable on a given file is limited by the ratio (file_size / data_size).

The following table shows the minimum uncompressed file size needed for full use of N processors at a given compression level, using the default data size for each level:

Processors 2 3 4 8 16 64
Level
-1 4 MiB 6 MiB 8 MiB 16 MiB 32 MiB 128 MiB
-2 6 MiB 9 MiB 12 MiB 24 MiB 48 MiB 192 MiB
-3 8 MiB 12 MiB 16 MiB 32 MiB 64 MiB 256 MiB
-4 12 MiB 18 MiB 24 MiB 48 MiB 96 MiB 384 MiB
-5 16 MiB 24 MiB 32 MiB 64 MiB 128 MiB 512 MiB
-6 32 MiB 48 MiB 64 MiB 128 MiB 256 MiB 1 GiB
-7 64 MiB 96 MiB 128 MiB 256 MiB 512 MiB 2 GiB
-8 96 MiB 144 MiB 192 MiB 384 MiB 768 MiB 3 GiB
-9 128 MiB 192 MiB 256 MiB 512 MiB 1 GiB 4 GiB


Next: , Previous: Minimum file sizes, Up: Top

7 Reporting bugs

There are probably bugs in plzip. There are certainly errors and omissions in this manual. If you report them, they will get fixed. If you don't, no one will ever know about them and they will remain unfixed for all eternity, if not longer.

If you find a bug in plzip, please send electronic mail to lzip-bug@nongnu.org. Include the version number, which you can find by running 'plzip --version'.


Previous: Problems, Up: Top

Concept index